Measurement

A form click is not a successful enquiry

Define form attempts, successful submissions and qualified leads separately, then test the cases that make a conversion report misleading.

Name the outcome you can prove. A click shows intent; a confirmed submission shows that the application accepted the request. Neither alone proves a qualified lead.

Start with the promise in the report

When a report labels every submit-button click as a lead, failed attempts can look like business results. A visitor may click with a required field missing, lose their connection, or retry after an unclear response. The interaction happened, but the intended outcome may not have happened.

Agree on three separate concepts before choosing an event name: an attempt, an accepted submission, and a qualified opportunity. The last usually needs a downstream business decision. Keep that distinction visible in the dashboard, even when the application cannot yet supply every stage.

Attach measurement to the confirmed outcome

For a successful-submission metric, identify the point where the application knows that the request was accepted. Ask the developer what the response actually means. A thank-you message that appears before acceptance is not sufficient evidence; a page that can be refreshed or opened directly also needs a counting rule.

Google Analytics supports events for specific interactions and shows events and parameters in Realtime and DebugView. Those tools help check collection. They do not decide whether your chosen trigger represents a valid business outcome.

  • Write the trigger and the evidence of success.
  • Document whether retries should create a new counted outcome.
  • Keep validation failures and request failures separate from successes.
  • Agree which test traffic is excluded from business reporting.

Test more than the happy path

Use a test environment and synthetic submissions. Record the user-visible result, application result and observed measurement for each scenario. If they disagree, investigate before trusting the report.

A small test matrix is more useful than a green indicator with no explanation. It also gives the next person a way to reproduce the check after the form changes.

  • Empty required field: no confirmed success.
  • Valid accepted request: the intended success is recorded.
  • Request failure: no success is reported before acceptance.
  • Double click or retry: inspect duplicates against the agreed rule.
  • Return to the confirmation page: verify whether another outcome is counted.

Publish the definition beside the number

Use a label such as “accepted enquiries” only when that is what the implementation measures. Add the reporting period and known limitations. If the current event is a contact interaction, call it that while the stronger measurement is built.

This is a testing method, not a report of a measured conversion lift. Its benefit is a clearer decision: you can distinguish a tracking issue, a form failure and a lead-quality problem instead of treating them as the same thing.

Sources & context